Inter-Frame Motion Correction in Whole-Body Direct Parametric Image Reconstruction

1. A computer-implemented method, comprising: receiving a nuclear imaging data set including a set of dynamic frames; generating at least one of a whole-body forward motion field or a whole-body inverse motion field for at least one frame in the set of dynamic frames; applying an iterative loop to update at least one parameter used in a direct parametric reconstruction, wherein the iterative loop includes: calculating frame emission images for the at least one frame; generating motion-corrected frame emission images based on the at least one whole-body forward motion field or a whole-body inverse motion field; and updating the at least one parameter by applying a linear or nonlinear fit to the motion-corrected frame emission images; and generating at least one parametric image based on the at least one parameter updated by the iterative loop.

2. The computer-implemented method of claim 1 , wherein generating at least one of a whole-body forward motion field or a whole-body inverse motion field for at least one frame in the set of frames comprises: selecting a reference frame from the set of frames; calculating the at least one of the whole-body forward motion field or the whole-body inverse motion field for each frame other than the reference frame in the set of frames; and outputting the at least one of the whole-body forward motion field or the whole-body inverse motion field for each frame in the set of frames.

3. The computer-implemented method of claim 1 , wherein calculating the motion-corrected frame emission image comprises applying an activity reconstruction.

4. The computer-implemented method of claim 3 , wherein the activity reconstruction is performed according to: f i + 1 ⁡ ( t ) = f i ⁡ ( t ) ⁢ 1 M t ⁡ ( P - 1 ⁡ ( 1 AN ⁡ ( t ) ) ) ⁢ M t ⁡ ( P - 1 ⁡ ( y ⁡ ( t ) P ⁡ ( M t - 1 ⁡ ( f ⁡ ( t ) ) + r ⁡ ( t ) ) + s ⁡ ( t ) ) ) where f is the frame emission image, t is an index for a current frame in the set of frames, y is measured emission data, P is a forward projection, P −1 is an inverse projection, M t is the whole-body forward motion field for the current frame t, M t −1 is the whole-body inverse motion field for the current frame t, r is an expected value for randoms, s is an expected value for scatter, A is an attenuation correction factor, and Nis a frame-dependent normalization factor.

6. The computer-implemented method of claim 1 , wherein the at least one of a whole-body forward motion field or a whole-body inverse motion field for at least one frame in the set of frames is generated using a plurality of histo-images generated by a direct histogram of the nuclear imaging data set.

7. The computer-implemented method of claim 6 , wherein the histo-images are filtered using a neural network trained using one or more ground-truth fully reconstructed images.

8. The computer-implemented method of claim 1 , wherein the at least one parameter comprises a metabolic uptake rate (Ki) and a distribution volume (DV).

9. The computer-implemented method of claim 8 , wherein direct parametric reconstruction comprises a direct Patlak reconstruction.

10. The computer-implemented method of claim 1 , wherein the at least one parameter comprises a slope and an intercept.

11. The computer-implemented method of claim 1 , wherein the nuclear imaging data set is selected from the group consisting of a PET imaging data set, a SPECT imaging data set, or a CT imaging data set.

12. The computer-implemented method of claim 1 , wherein generating a motion-corrected frame emission image based on the at least one whole-body forward motion field or the whole-body inverse motion field comprises: applying the whole-body inverse motion field to the frame emission image; calculating a correction image; and applying the whole-body forward motion field to the correction image.

13. The computer-implemented method of claim 12 , wherein generating a motion-corrected frame emission image based on the at least one whole-body forward motion field or the whole-body inverse motion field further comprises dividing correction image by at least one forward warped normalization factor.

14. A system, comprising: a nuclear imaging scanner configured to obtain a set of nuclear imaging data including a set of frames; and a processor configured to: receive the nuclear imaging data set from the nuclear imaging scanner; generate at least one of a whole-body forward motion field or a whole-body inverse motion field for at least one frame in the set of frames; apply an iterative loop to update at least one parameter used in a direct parametric reconstruction, wherein the iterative loop includes: calculating a frame emission image for the at least one frame; generating a motion-corrected frame emission image based on the at least one whole-body forward motion field or a whole-body inverse motion field; and updating the at least one parameter by applying a linear or nonlinear fit to the motion-corrected frame emission image; and generate at least one parametric image based on the at least one parameter updated by the iterative loop.

15. The system of claim 14 , wherein the motion-corrected frame emission image is calculated by applying a parametric activity reconstruction according to: f i + 1 ⁡ ( t ) = f i ⁡ ( t ) ⁢ 1 M t ⁡ ( P - 1 ⁡ ( 1 AN ⁡ ( t ) ) ) ⁢ M t ⁡ ( P - 1 ⁡ ( y ⁡ ( t ) P ⁡ ( M t - 1 ⁡ ( f ⁡ ( t ) ) + r ⁡ ( t ) ) + s ⁡ ( t ) ) ) where f is the frame emission image, t is an index for a current frame in the set of frames, y is measured emission data, P is a forward projection, P −1 is an inverse projection, M t is the whole-body forward motion field for the current frame t, M t −1 is the whole-body inverse motion field for the current frame t, r is an expected value for randoms, s is an expected value for scatter, A is an attenuation correction factor, and Nis a frame-dependent normalization factor.

17. The system of claim 14 , wherein the at least one parameter comprises a metabolic uptake rate (Ki) and a distribution volume (DV).

19. A non-transitory computer readable medium storing instructions configured to cause a computer system to execute the steps of: receiving a nuclear imaging data set including a set of frames; generating at least one of a whole-body forward motion field or a whole-body inverse motion field for at least one frame in the set of frames; applying an iterative loop to update at least one parameter used in a direct parametric reconstruction, wherein the iterative loop includes: calculating a frame emission image for the at least one frame; generating a motion-corrected frame emission image based on the at least one whole-body forward motion field or a whole-body inverse motion field; and updating the at least one parameter by applying a linear fit to the motion-corrected frame emission image; and generating at least one parametric image based on the at least one parameter updated by the iterative loop.

20. The non-transitory computer readable medium of claim 19 , wherein calculating the motion-corrected frame emission image comprises applying a parametric activity reconstruction according to: f i + 1 ⁡ ( t ) = f i ⁡ ( t ) ⁢ 1 M t ⁡ ( P - 1 ⁡ ( 1 AN ⁡ ( t ) ) ) ⁢ M t ⁡ ( P - 1 ⁡ ( y ⁡ ( t ) P ⁡ ( M t - 1 ⁡ ( f ⁡ ( t ) ) + r ⁡ ( t ) ) + s ⁡ ( t ) ) ) where f is the frame emission image, t is an index for a current frame in the set of frames, y is measured emission data, P is a forward projection, P −1 is an inverse projection, M t is the whole-body forward motion field for the current frame t, M t −1 is the whole-body inverse motion field for the current frame t, r is an expected value for randoms, s is an expected value for scatter, A is an attenuation correction factor, and Nis a frame-dependent normalization factor.